Integrating Artificial Intelligence For Early Prediction Of Preterm Labor Using Cardiotocography And Cervical Length: A Machine Learning-Based Clinical Decision Support Approach

Abstract

Preterm labor, a leading cause of neonatal mortality and morbidity, remains a significant obstetric challenge worldwide. Early identification of at-risk pregnancies can dramatically improve outcomes through timely interventions. Traditional diagnostic methods such as cardiotocography (CTG) and transvaginal cervical length assessment have been individually utilized to estimate preterm birth risk but lack integrated predictive power. This study aims to harness the capabilities of artificial intelligence (AI) by developing a machine learning (ML)-based predictive model that combines CTG and cervical length data to forecast preterm labor risk. A retrospective dataset comprising 800 cases from tertiary care centers was analyzed. Multiple algorithms—Random Forest, Support Vector Machine (SVM), XG Boost, and Artificial Neural Networks (ANNs)—were trained and validated using cross-validation techniques. The final model demonstrated an AUC of 0.91, suggesting high diagnostic value. Integration of AI in obstetrics presents a paradigm shift in prenatal care, enabling earlier detection and individualized risk management of preterm labor.